Author(s): Victor Dupuis; Sebastien Proust; Celine Berni; Andre Paquier; Fabien Thollet
Linked Author(s): André Paquier, Sébastien Proust, Celine Berni
Keywords: Vegetation; Roughness transition; Laboratory study; Floodplain; Open-channel flow
Abstract: The understanding of vegetated flows is of primary importance since a new trend in river management consists in restoring rivers and floodplains to their natural form. The present laboratory study focuses on the particular case of a floodplain with a longitudinal roughness step change from highly-submerged dense meadow to emergent rigid stems set on the dense meadow (representing a wood) ,and vice versa. First, it was found that the flow depth solely varies upstream from the roughness change (but the flume is too short to observe the uniform flow depth at the upstream end of the flume) .In this region, the vertical profiles of mean velocity, Reynolds stress and turbulent intensities are self-similar, when the mean streamwise velocity is normalized by the bulk velocity and the turbulent quantities by the shear velocity. Downstream from the roughness step change, the mean flow and turbulent quantities are spatially evolving over longitudinal distances that are about 35 to 50 times the water depth. Second, a simple 1D model is found to accurately predict the longitudinal profile of water depth through the two types of roughness transitions (from meadow to rigid stems and vice versa) .Bed-roughness is modelled by Manning’s formula and the drag caused by the stems is modelled by a volume force with constant drag coefficient and constant frontal surface per unit area. Results of this model show that the convergence length upstream of the roughness step change, that is necessary to reach the uniform water depth of the upstream roughness, scales with the uniform water depth of the upstream roughness Hup and can be estimated in the present case as 2000 Hup.